AMSA Expanding Anti-Industry Scorecard to 400 Teaching Hospitals
We recently reported on a survey of 1,610 first- and third-year medical students and 739 residents regarding conflict of interest (COI) policies and their interactions with industry. In our story, we noted that almost every year, the American Medical Students Association (AMSA) publishes a yearly “PharmFree Scorecard” that evaluates the COI policies at American medical schools. Here is a little background and history on AMSA.
The PharmFree Scorecard is funded through the Consumer and Prescriber Education grant program, which resulted from the Neurontin settlement back in 2004.
It is because of organizations like PharmFree that frame all physician-industry collaboration as negative or unethical, that patients, physicians and biopharmaceutical professionals need a voice in the “conflict of interest mania,” such as the recently launched Partners for a Health Dialogue.
Interestingly, while we cited to the 2012 results, AMSA just released its 2013 Scorecard, the sixth iteration of the scorecard. The release comes as AMSA hosts its National PharmFree Week April 8-12, 2013. The annual event “highlights the importance of putting patients first by addressing conflicts of interest and encouraging evidence rather than marketing-based education.” Below is a summary of this year’s findings as well as additional information about AMSA activities.
Interestingly, AMSA and Pew announced with this year’s data that they were also charged through a grant from the Oregon State Attorney General’s office to update the AMSA PharmFree Scorecard and expand it to 400 teaching hospitals. Alongside Dr. Joseph Ross of Yale Medical School who served as a Methodology Consultant, AMSA and Pew developed the new AMSA PharmFree Scorecard set for release in 2014. Given that payments to teaching hospitals must be reported under the Physician Payment Sunshine Act, the new scoring of teaching hospitals will bring increased scrutiny to physician-industry interactions at such hospitals.
There will be two versions of the new Scorecard - one for medical schools and another for teaching hospitals. The number of domains for the Scorecard has also expanded with 16 domains common to both versions and 2 additional domains for “Samples” and “Purchasing and Formulary Committees” specific to the teaching hospital version. Other revisions to the Scorecard will include a new formula for assessing academic medical centers as well as revisions in the framework for grading schools.
“Conflict of interest policies discourage inappropriate relationships and allow transparent and positive industry-physician collaborations to thrive,” said Daniel Carlat, director of the Pew Charitable Trusts’ prescription project, which supports AMSA’s work. “The 2013 scorecard shows medical schools are moving toward stronger conflict-of-interest standards.”
As of April 9, 2013, 153 out of 158 medical institutions considered eligible for grading have participated in the Scorecard, a 97% participation rate. Of these 158 US medical schools, 40 receive “A”s (25%), 75 “B”s (47%), 14 “C”s (9%), and 13 “D”s (8%). Thus, 115 of 158 medical schools (73%) now receive a grade of A or B for their COI policies, compared with 102 last year. Approximately 8% of U.S. medical schools improved their COI policies since the 2011-2012 Scorecard.
Policies that apply only to medical students and/or residents, but not faculty, and (2) Policies that are only guidelines, without formal requirements, may only achieve a maximum grade of C.
13 schools (8%) receive a grade of F. This includes 3 schools that submitted policies graded as F, two schools that stated they had no COI policy in place, and five schools that did not respond to repeated attempts at follow-up in 2008, 2009, 2010, 2011-12 and 2012-2013. Three additional schools received an F as they did not submit new policies or demonstrate a continuing policy development process after remaining In Process for one year. Three schools (2%) received a grade of In Process. Here are some highlights from this year’s report:
- Roughly 80% of medical schools have perfect or close to perfect policies for on-campus CME.
- Only 4 medical schools – Univ. of South Dakota Sanford School of Medicine, Florida State Univ. College of Medicine, Stony Brook Univ. School of Medicine, and the Commonwealth Medical College -- completely ban sales representatives from campus.
- Only 41 schools (26%) have model policy in terms of disclosure, requiring personnel to disclose past and present financial ties with industry (e.g., consulting and speaking agreements, research grants) on a publicly-available website and disclosing these relationships to patients.
- All 8 of the Texas medical schools scored at least a B. The University of Texas Medical Branch at Galveston, however, remains the lone A in the Lone Star state. All three of the Maryland medical schools scored at least a B, with both Johns Hopkins and University of Maryland Schools of Medicine receiving an A.
- Schools with model policies on speaking arrangements have grown tremendously; 44 schools ban or severely restrict participation in speaker bureaus.
Specifically, 18 schools have banned participation by their faculty on speakers’ bureaus: Harvard Medical School, NYU School of Medicine, Duke Univ. School of Medicine, Columbia Univ. College of Physicians and Surgeons, Univ. of Arkansas School of Medicine, Univ. of Maryland School of Medicine, Georgia Health Sciences Univ., Univ. of South Carolina School of Medicine, Univ. of Hawai’i John A Burns School of Medicine, Creighton Univ. School of Medicine, Wake Forest Univ. School of Medicine, Univ. of Massachusetts Medical School, Emory Univ. School of Medicine, Stanford Univ. School of Medicine, Albert Einstein College of Medicine, Univ. of Alabama Birmingham, Univ. of Florida and Jefferson Medical College.
The following is a list of domains and the number of perfect scores in each in 2013:
- 28 Schools have perfect scores for “On-campus continuing medical education;” improved from 20 in 2011-2, 15 in 2010, 3 in 2009 and 5 in 2008. On-site education means within the medical school or hospital campus. To gain a perfect score, “Industry is not permitted to provide direct financial support for educational activities, including CME, directly or through a subsidiary agency. (However, companies may contribute unrestricted funds to a central fund or oversight body at the AMC, which, in turn, would pool and disburse funds for programs that are independent of any industry input or control.)”
- 98 medical schools have “Less stringent limitations to ensure independence of educational content (e.g., standards to establish freedom from industry influence of content, such as review and approval of presentations; language that prevents industry from selecting the speaker; a requirement that programs adhere to ACCME standards; or language such as: industry funding may be allocated for a particular topic, but must be provided directly to the department, not to individuals). Despite trying to discredit the ACCME as a “non-governmental” organization, CMS recently recognized the critical importance of the ACCME Standards for Commercial Support in the final Sunshine Act rule, and non-compliance with such standards can result in loss of accreditation.
- Off-campus continuing medical education – 102; improved from 88 in 2011-2, 75 in 2010, 49 in 2009 and 23 in 2008. Off-site education is at outside facilities, including professional conferences. AMSA measures this category based on how schools regulate compensation for travel or attendance at off-site lectures and meetings. To gain a perfect score, “Personnel may not accept payment, gifts or financial support from industry to attend lectures and meetings. (An exception may be made for modest meals, if part of a larger program.) Travel support may only be accepted if it is subject to institutional approval or industry is prevented from selecting (“earmarking”) the recipients.” It is unclear, however, whether this category captures accredited CME or simply all off-campus educational events.
- Scholarships & Funds for Trainees – 123; improved from 108 in 2011-2, 94 in 2010, 66 in 2009 and 29 in 2008. To gain a perfect score, “The policy must either prevent industry from earmarking or awarding funds to support the training of particular individuals (recipients must be chosen by the school or department), or the policy must mandate institutional review of the giving of funds. (This does not preclude grants that fund a specific research project.)”
- Gifts & Meals – 93; improved from 81 in 2011-2, 66 in 2010, 44 in 2009 and 19 in 2008. To gain a perfect score in Gifts & meals, “All gifts and on-site meals funded by industry are prohibited, regardless of nature or value.” Less stringent include limitations such as prohibitions above $50 per year, or gifts prohibited but meals allowed.
- Disclosure – 41; improved from 29 in 2011-2, 20 in 2010, 5 in 2009 and 1 in 2008. To gain a perfect score, “Personnel are required to disclose past and present financial ties with industry (e.g., consulting and speaking agreements, research grants) on a publicly-available website and/or disclose such relationships to patients when such a relationship might represent an apparent conflict of interest.”
- Curriculum – 79; improved from 69 in 2011-2, 48 in 2010, 28 in 2009, and 12 in 2008. To gain a perfect score, “Students are trained to understand institutional conflict-of-interest policies and recognize how industry promotion can influence clinical judgment.”
- Consulting (excluding scientific research and speaking) – 71, improved from 64 in 2011-2, 49 in 2010, 27 in 2009, and 12 in 2008. To gain a perfect score, consulting relationships “must be subjected to institutional review or approval. Additionally, they must either be described in a formal contract, or payment for services must be commensurate to the task.”
- Sales representatives – 4; improved from both 2 in 2011-2 and 2010, and both 1 in 2009 and 2008. To gain a perfect score, “Pharmaceutical and device representatives are not allowed to meet with faculty regardless of location, or are not permitted to market their products anywhere inside the medical center and associated clinics and offices. (Exceptions may be made for non-marketing purposes, such as training on devices or equipment.)”
- Purchasing & Formularies – 83; improved from 70 in 2011-2, 66 in 2010, 47 in 2009, and 22 in 2008. To gain a perfect score, “Formulary committees and committees overseeing purchases of medical devices should exclude those who have financial relationships with drug or device manufacturers. Exclusion may be specific to participation in particular decisions for which the staff member has a conflict of interest. This policy does not prevent expert clinicians from advising a committee, provided that potential conflicts are disclosed. (Note: this standard is not intended to prohibit indirect financial interests, such as investments in mutual funds that may own pharmaceutical company shares).”
- Samples – 42, improved from 31 in 2011-2, 26 in 2010, 20 in 2009, and 12 in 2008. To gain a perfect score, “Industry samples are prohibited, except under certain narrow circumstances approved by the institution that protect the interests of patients and prevent the use of samples as a marketing tool (e.g., policies that allow samples under limited circumstances with the approval of the Pharmacy and Therapeutics (P&T) Committee or policies that incorporate samples into a larger program designed to ensure the availability of brand-name and generic medications to under-insured patients; if the circumstances of the specific program are not defined, the policy should define the approvals process). Where there is a specific program in place, the policy must prevent samples from being given directly to physicians by pharmaceutical sales representatives.”
- Industry-funded speaking – 44; improved from 31 in 2011-2, 19 in 2010, 10 in 2009, and 4 in 2008. To gain a perfect score, “Speaking relationships are prevented from functioning as de facto gifts or marketing. An effective policy must not implicitly permit (a) long-term speaking agreements or (b) industry to have a role in determining presentation content. (Some effective policies may explicitly prohibit participation in a speakers bureau. Other effective policies contain elements such as limits on compensation and reimbursement and a requirement to ensure the scientific integrity of information presented.)”
Similar to the results of the past, the areas that garnered the greatest number of perfect scores were those addressing industry support of scholarships, off-campus “CME,” faculty participation in industry-speaking relationships, purchasing and formulary committees, and gifts. Industry support of scholarships experienced the largest net growth in achieving perfect scores in this domain (15 additional medical schools). New areas being addressed by schools include samples as well as purchasing and formula committees. AMSA says that “on-campus CME” remains a challenging area.
Californian medical schools continued to excel with 7 of their 10 schools receiving an A grade. This is in part due to the strong system-wide University of California (UC) COI policy, which when graded on its own receives an A grade. However, some of the UC schools go even further by supplementing the system-wide policy with their own regulations. Texan schools continue to show significant improvement. All eight Texan medical schools score at least a B. University of Texas Medical Branch at Galveston, however, remains the lone A in the Lone Star state.
Another major training ground, Massachusetts, continues to show case schools with model policies. This year, University of Massachusetts Medical School and Boston University both received a B grade while Harvard Medical School and Tufts University School of Medicine both received an A grade.
The nine Pennsylvania medical schools have strong policies and in Maryland, all three schools (Johns Hopkins University, University of Maryland, and Uniformed Services of the Health University) received a B grade or higher. Finally, the state of Florida continues to perform well.
The PharmFree Scorecard methodology was developed jointly by AMSA and the Pew Prescription Project, an initiative of the Pew Health Group. Assessed domains are broadly consistent with those identified in recent literature – primarily Brennan et al. Health Industry Practices that Create Conflicts of Interest: A Policy Proposal for Academic Medical Centers. JAMA 2006; 295(4): 429-433.
The Scorecard assesses policies related to: (1) acceptance of gifts and meals from industry; (2) consulting relationships; (3) speaking relationships; (4) disclosure of financial conflicts; (5) pharmaceutical samples; (6) individuals with financial conflicts participating in university purchasing decisions; (7) financial support for educational events (on- and off-campus); (8) industry support for scholarships and trainee funds; (9) access of industry sales personnel to medical school or hospital personnel; and (10) inclusion of education about conflict of interest within the academic curriculum. Additionally, the presence of oversight and sanctions is examined, but not included in grade calculation.
AMSA maintains that “Two blinded assessors independently score each set of policies in the eleven areas included in the scorecard.” How can such assessors be blinded if they are working for AMSA/Pew, and are clearly anti-industry and biased towards any kind of industry interactions or collaboration?
Other AMSA Initiatives
Of particular interest is a new initiative known as the “2nd Slide Campaign,” which is based off the standard practice in graduate and continuing medical education (CME) to disclose conflicts of interest on the 2nd slide of one’s presentation. “The premise of this campaign is simple: medical students should receive the same types of disclosures that educators provide in other types of medication.” AMSA notes that “most medical schools lack clear guidelines to help faculty members disclose conflicts of interest to students.”
AMSA provides a sample disclosure slides and other information for medical students to bring about change in their medical schools. AMSA provides three scenarios for when a lecturer should disclose interests to students, and tailors their example disclosure slides to such scenarios:
- Lecturer has conflicts of interest but they are not relevant
- The lecturer has no conflicts of interest
- The lecturer has relevant conflicts of interest to disclose to the students
AMSA said that such disclosures matter because it is “vital” that medical students and “know where information comes from.” AMSA calls for such disclosure because “research shows that financial conflicts of interest can impact (even subconsciously!)the presentation and interpretation of data.” AMSA also has resourcs for a PharmFree Curriculum and “Best Practice Policies,” such as for gifts, meals, and samples.
Given the already hurried schedules of physicians and medical students, will there really be enough time for medical professors to address and discuss potential conflicts with students? Will such disclosures even matter to students? It seems more appropriate that this subject be clearly addressed in a standard medical ethics course over the course of medical school, not during each lecture given, particularly in light of the Sunshine Act, which will allow medical students to do their own research on professors, if they desire.
One interesting new program that AMSA appears to have started is the “National Opt Out Day.” Specifically, AMSA “asked graduating fourth year members to tell the American Medical Association (AMA) that they are opting out of the AMA Physician Masterfile!” “As physicians who strive to practice evidence-based medicine, we do not want our personal and prescribing information sold to the pharmaceutical and medical device industries.” The AMA Physician Masterfile was established by the AMA in 1906 as a record keeping device supporting membership and mailing activities. The Physician Masterfile includes current and historical data for more than 1.4 million physicians, residents and medical students in the United States.
AMSA also hosted a webinar with the University of Miami Miller School of Medicine to discuss how the school improved their conflict of interest policies as well as the barriers both current and past to implementing change. Executive Dean of Education and Policy Dr. Laurence Gardner will be presenting.
AMSA noted that in addition to Pew, it has joined forces with the National Physicians Alliance (NPA) and Community Catalyst to create both external and internal pressure for medical schools and academic medical centers to adopt strong COI policies, through a new initiative- the Partnership to Advance Conflict-Free Medical Education (PACME). They noted how NPA’s Unbranded Doctor campaign provides resources for conflict-free medical practice including an archive of past events including bimonthly Conflict-Free Leadership Calls and National Grand Rounds.
Community Catalyst is in the process of developing a series of toolkits to help institutions improve their policies with examples of language from other model institutions and provides technical assistance to schools via regional roundtable and individual consultation.